ksm26 / langchain-for-llm-application-development

Apply LLMs to your data, build personal assistants, and expand your use of LLMs with agents, chains, and memories.

🚀 LangChain-for-LLM-Application-Development

💻 Welcome to the "LangChain for LLM Application Development" course! This one-hour course, instructed by Harrison Chase, the creator of the LangChain framework, along with Andrew Ng, will equip you with essential skills to enhance the use cases and capabilities of language models in application development.

LangChain: 🔗GitHub, 📚Documentation

Course Summary

In this course, you will dive into various topics that will empower you to leverage the LangChain framework effectively. Here's what you can expect to learn and experience:

1. 📚 Models, Prompts and Parsers: Discover how to call Language Models (LLMs), provide prompts, and parse the responses.
2. 🧠 Memories for LLMs: Learn how to employ memories to store conversations and manage limited context space effectively.
3. ⛓️ Chains: Explore the creation of sequences of operations using LangChain.

4. 📄 Question Answering over Documents: Apply LLMs to your proprietary data and tailor them to meet specific use case requirements.

5. 👥 Agents: Gain insights into the emerging development of LLMs as reasoning agents.

By the end of this course, you will have a solid foundation in using LangChain and will be equipped with a model that can serve as a starting point for developing your own applications using diffusion models.

Key Points

• 🔑 Learn LangChain directly from Harrison Chase, the creator of the framework.
• 🤖 Apply LLMs to your proprietary data and develop personalized assistants and specialized chatbots.
• 💡 Expand your utilization of LLMs through agents, chained calls, and memories.

About the Instructors

🌟Harrison Chase is Co-Founder and CEO at LangChain.

🌟Andrew Ng is a renowned AI researcher, co-founder of Coursera, and the founder of DeepLearning.AI. With a wealth of knowledge and expertise in the field, Andrew has played a pivotal role in popularizing AI education.

🔗"LangChain for LLM Application Development" course. To enroll in the course or for further information, visit deeplearning.ai.

langchain-for-llm-application-development's Issues

L3-chains.ipynb: OutputParserException in line 34

I got the following error when running L3-chains.ipynb, line 34

chain.run("What is black body radiation?")

OutputParserException: Parsing text
{
    "destination": "physics",
    "next_inputs": "What is black body radiation?"
}
 raised following error:
Got invalid return object. Expected markdown code snippet with JSON object, but got:
{
    "destination": "physics",
    "next_inputs": "What is black body radiation?"
}

I think it's the prompt issue that GPT can not understand the following prompt line 31:

MULTI_PROMPT_ROUTER_TEMPLATE = """Given a raw text input to a \
...
...
<< OUTPUT (remember to include the ```json)>>"""

So it forget to wrap the output with ```json

I suggest use the following prompt instead which is more explicit

<< OUTPUT (remember to wrap the output with ```json (output)```)>>"""

is:issue is:open Getting UnicodeDecodeError Error When Running VectorstoreIndexCreator

I'm getting the following error when implementing this locally:

UnicodeDecodeError                        Traceback (most recent call last)
Cell In[56], line 7
      3 file = 'C:/Users/rasha/OneDrive/Documents/Projects/ttdash/OutdoorClothingCatalog_1000.csv'
      4 loader = CSVLoader(file_path=file)
      6 index = VectorstoreIndexCreator(
----> 7 vectorstore_cls=DocArrayInMemorySearch).from_loaders([loader])
      8 response = index.query(prompt)

File ~\miniconda3\envs\ttdash_env\Lib\site-packages\langchain\indexes\vectorstore.py:72, in VectorstoreIndexCreator.from_loaders(self, loaders)
     70 docs = []
     71 for loader in loaders:
---> 72     docs.extend(loader.load())
     73 return self.from_documents(docs)

File ~\miniconda3\envs\ttdash_env\Lib\site-packages\langchain\document_loaders\csv_loader.py:47, in CSVLoader.load(self)
     45 with open(self.file_path, newline="", encoding=self.encoding) as csvfile:
     46     csv_reader = csv.DictReader(csvfile, **self.csv_args)  # type: ignore
---> 47     for i, row in enumerate(csv_reader):
     48         content = "\n".join(f"{k.strip()}: {v.strip()}" for k, v in row.items())
     49         try:

File ~\miniconda3\envs\ttdash_env\Lib\csv.py:111, in DictReader.__next__(self)
    108 if self.line_num == 0:
    109     # Used only for its side effect.
    110     self.fieldnames
--> 111 row = next(self.reader)
    112 self.line_num = self.reader.line_num
    114 # unlike the basic reader, we prefer not to return blanks,
    115 # because we will typically wind up with a dict full of None
    116 # values

File ~\miniconda3\envs\ttdash_env\Lib\encodings\cp1252.py:23, in IncrementalDecoder.decode(self, input, final)
     22 def decode(self, input, final=False):
---> 23     return codecs.charmap_decode(input,self.errors,decoding_table)[0]

UnicodeDecodeError: 'charmap' codec can't decode byte 0x9d in position 4594: character maps to <undefined>

Any reason why even though I'm using the same CSV data used in the original notebook?